Granted Patent B2

US 12,657,399 · App. 18/419,844 · Granted Jun 16, 2026

Inventors: Scott Pitkin (Bloomington, IN); Michael Aurelio (Novato, CA); Jonathan Belkowitz (Brooklyn, NY); Allen Hoem (Indianapolis, IN); Amrutha Krishnan (San Bruno, CA); Abigail Kutruff (New York, NY)

Assignee: Salesforce, Inc.

G06F40/40G06Q30/0243

Abstract

Claims (35)

1 . A computing services environment including a hardware processor and memory, the computing services environment comprising:

a database system including one or more database tables storing a plurality of database records reflecting customer relations management interactions between a plurality of tenants of the computer services environment and individuals interacting with the plurality of tenants, the database system configured to determine an input description of a communication campaign between a tenant of the plurality of tenants and a corresponding segment of the individuals;

a generative language model interface configured to determine a textual description of one or more elements of the communication campaign by completing a campaign brief generation prompt via a generative language model, the campaign brief generation prompt including some or all of the input description and a natural language instruction executed by the generative language model to generate the textual description, the generative language model interface being configured to determine a plurality of alternative textual element descriptions for an element of the one or more elements;

a communication interface configured to transmit to a client machine authenticated to a database system account linked to the tenant an instruction to generate a graphical user interface at the client machine, the graphical user interface identifying the segment of the individuals and the textual description, the graphical user interface receiving user input indicating a refinement to the textual description;

a messaging interface configured to transmit messages to selected individuals from the segment of individuals based on the input description and the refinement, the messaging interface being configured to transmit a first message including a first textual element description of the plurality of alternative textual element descriptions to a first subset of the segment of individuals and to transmit a second message including a second textual element description of the plurality of alternative textual element descriptions to a second subset of the segment of individuals; and

updating the communication campaign via the generative language model based on a performance metric determined based on the messages.

2 . The computing services environment recited in claim 1 , wherein the input description of the communication campaign identifies a product or service provided by the tenant.

3 . The computing services environment recited in claim 1 , wherein the computing services environment is configured to determine a first campaign performance metric for the first message and a second campaign performance metric for the second message, and wherein the computing services environment is configured to select the first message for transmission to a third subset of the segment of individuals based at least in part on a comparison of the first campaign performance metric and the second campaign performance metric.

4 . The computing services environment recited in claim 1 , wherein the one or more elements include a description of the segment of the individuals.

5 . The computing services environment recited in claim 1 , wherein the one or more elements include messaging text to include in one or more of the messages.

6 . The computing services environment recited in claim 5 , wherein the one or more elements include a description of a communication tone for generating the messaging text.

7 . The computing services environment recited in claim 5 , wherein the one or more elements include a performance indicator for the messages.

8 . The computing services environment recited in claim 1 , wherein the campaign brief generation prompt is determined by replacing a fillable portion of a campaign brief generation prompt template with some or all of the input description, the campaign brief generation prompt template including the natural language instruction.

9 . A method implemented at a computing services environment, the method comprising:

storing a plurality of database records in one or more database tables of a database system, the plurality of database records reflecting customer relations management interactions between a plurality of tenants of the computer services environment and individuals interacting with the plurality of tenants;

determining via a processor an input description of a communication campaign between a tenant of the plurality of tenants and a corresponding segment of the individuals;

determining a textual description of one or more elements of the communication campaign by completing a campaign brief generation prompt via a generative language model, the campaign brief generation prompt including some or all of the input description and a natural language instruction executed by the generative language model to generate the textual description, the generative language model interface being configured to determine a plurality of alternative textual element descriptions for an element of the one or more elements;

transmitting to a client machine authenticated to a database system account linked to the tenant an instruction to generate a graphical user interface at the client machine, the graphical user interface identifying the segment of the individuals and the textual description;

receiving via a communication interface user input provided via the graphical user interface receiving, the user input indicating a refinement to the textual description;

transmitting messages to selected individuals from the segment of individuals based on the input description and the refinement via a messaging interface, the messaging interface being configured to transmit a first message including a first textual element description of the plurality of alternative textual element descriptions to a first subset of the segment of individuals and to transmit a second message including a second textual element description of the plurality of alternative textual element descriptions to a second subset of the segment of individuals; and

updating the communication campaign via the generative language model based on a performance metric determined based on the messages.

10 . The method recited in claim 9 , wherein the input description of the communication campaign identifies a product or service provided by the tenant.

11 . The method recited in claim 9 , wherein the computing services environment is configured to determine a first campaign performance metric for the first message and a second campaign performance metric for the second message, and wherein the computing services environment is configured to select the first message for transmission to a third subset of the segment of individuals based at least in part on a comparison of the first campaign performance metric and the second campaign performance metric.

12 . The method recited in claim 9 , wherein the one or more elements include a description of the segment of the individuals.

13 . The method recited in claim 9 , wherein the one or more elements include messaging text to include in one or more of the messages.

14 . The method recited in claim 13 , wherein the one or more elements include a description of a communication tone for generating the messaging text.

15 . The method recited in claim 9 , wherein the campaign brief generation prompt is determined by replacing a fillable portion of a campaign brief generation prompt template with some or all of the input description, the campaign brief generation prompt template including the natural language instruction.

16 . One or more non-transitory computer readable media having instructions stored thereon for performing a method implemented at a computing services environment, the method comprising:

storing a plurality of database records in one or more database tables of a database system, the plurality of database records reflecting customer relations management interactions between a plurality of tenants of the computer services environment and individuals interacting with the plurality of tenants;

determining via a processor an input description of a communication campaign between a tenant of the plurality of tenants and a corresponding segment of the individuals;

determining a textual description of one or more elements of the communication campaign by completing a campaign brief generation prompt via a generative language model, the campaign brief generation prompt including some or all of the input description and a natural language instruction executed by the generative language model to generate the textual description, the generative language model interface being configured to determine a plurality of alternative textual element descriptions for an element of the one or more elements;

transmitting to a client machine authenticated to a database system account linked to the tenant an instruction to generate a graphical user interface at the client machine, the graphical user interface identifying the segment of the individuals and the textual description;

receiving via a communication interface user input provided via the graphical user interface receiving, the user input indicating a refinement to the textual description;

transmitting messages to selected individuals from the segment of individuals based on the input description and the refinement via a messaging interface, the messaging interface being configured to transmit a first message including a first textual element description of the plurality of alternative textual element descriptions to a first subset of the segment of individuals and to transmit a second message including a second textual element description of the plurality of alternative textual element descriptions to a second subset of the segment of individuals; and

updating the communication campaign via the generative language model based on a performance metric determined based on the messages.

Assignments (1)

Continuity (3)

References Cited (20)